The Broad Transcription Factor Links Hormonal Signaling, Gene Expression, and Cellular Morphogenesis Events During Drosophila Imaginal Disc Development [Developmental and Behavioral Genetics]

TISSUE morphogenesis is required for the elaboration of the body axis and organs during metazoan development. In some contexts, hormonal signals provide temporal cues to coordinate these morphogenetic events. Imaginal disc morphogenesis in the fruit fly Drosophila melanogaster provides an excellent model to uncover molecular mechanisms by which hormonal signals are translated into the physical changes that occur during development. Imaginal discs are diploid tissues found within the larva that give rise to the adult integument during metamorphosis (Willis 1974; Csaba 1977; Fristrom 1988). Both classical experiments and more recent live imaging studies (De las Heras et al. 2018; Diaz-de-la-Loza et al. 2018) have revealed requirements for cell shape changes and rearrangements, as well as for remodeling of the extracellular matrix (ECM) in imaginal disc morphogenesis during metamorphosis. These experiments also demonstrate the key role that 20-hydroxyecdysone (hereafter referred to as ecdysone) plays in directing these processes. Ecdysone acts through a transcriptional cascade: the hormone binds to its heterodimeric receptor, which acts as a DNA-binding activator to drive transcription of early-response genes, including Eip74EF, Eip75B, and broad (br), which themselves encode DNA-binding proteins that activate late-response genes (Chao and Guild 1986; Feigl et al. 1989; Janknecht et al. 1989; Burtis et al. 1990; Segraves and Hogness 1990; DiBello et al. 1991; Yao et al. 1993; Crossgrove et al. 1996). Despite extensive research into Drosophila metamorphosis, parts of the pathway connecting the ecdysone cue to the effectors involved in imaginal disc morphogenesis have yet to be elucidated.

Of the ecdysone signaling early-response genes, br appears to have the most direct effects on imaginal disc morphogenesis (Kiss et al. 1988). br encodes four transcription factors, each carrying a unique zinc finger domain spliced to a common Broad-Complex, Tramtrack, and Bric a Brac/Pox virus and Zinc finger (BTB/POZ) domain (DiBello et al. 1991; von Kalm et al. 1994; Bayer et al. 1996) (Supplemental Material, Figure S1). These four isoforms, known as the Z1, Z2, Z3, and Z4 isoforms, have three genetically separate functions (br, reduced bristles on the palpus, and 2Bc), with the Z2 isoform performing the classic br function (Bayer et al. 1997). This function is critical for imaginal disc morphogenesis: animals lacking functional Br Z2 have discs that fail to elongate, and these animals die during the prepupal stage before head eversion (Kiss et al. 1988). The various isoforms of Br also add a layer of complexity to the regulation of metamorphic genes. br does not simply respond to the ecdysone cue through a general increase in transcription that subsequently effects the increased transcription of late-response genes; rather, the isoforms exhibit a dynamic pattern of expression that differs by tissue (Huet et al. 1993). In the imaginal discs, the expression of the Z2 transcript rises dramatically approximately 4 hr before pupariation, but greatly decreases in the hours after pupariation, while the expression of the Z1 transcript greatly increases between 2 and 4 hr after pupariation (Emery et al. 1994; Bayer et al. 1996). This late larval pulse of Z2 expression is critical for activation of late-response genes; however, the identities of these genes remain largely unknown. The illumination of the link between the ecdysone cue and morphogenetic effects in imaginal discs hinges upon the identification of these late-response genes and how they are regulated by ecdysone and br.

Previous screens using the hypomorphic br1 allele identified a number of br-interacting genes, including Stubble (Sb), zipper, Rho1, Tropomyosin 1, blistered, and ImpE3 (Beaton et al. 1988; Ward et al. 2003), although the majority of these genes are not transcriptional targets of br (Ward et al. 2003). Nevertheless, the identity of these genes suggests roles for br in the major processes implicated in hormonal control of imaginal disc morphogenesis, namely cell shape changes/rearrangements and modification of the ECM. Here, we show that cell shape changes and cell rearrangements fail to occur normally in br5 mutant prepupae. In addition, consistent with previous work demonstrating that the ECM provides a constraining force and must be degraded to allow disc elongation (Pastor-Pareja and Xu 2011; Diaz-de-la-Loza et al. 2018), we show that the basal ECM protein Collagen IV is not substantially degraded in leg imaginal discs from amorphic br5 mutant animals as old as 8 hr after puparium formation (APF). In this study, we also use this allele to identify specific genes regulated by br in the leg discs at the onset of metamorphosis through an RNA-sequencing-based approach. This approach identified over 700 br-regulated genes, including genes with known metabolic and developmental functions, including ECM organization and modification. We tested a subset of these genes for roles in leg morphogenesis through RNA interference (RNAi) and found several that are necessary for proper development of the adult legs. These results demonstrate the value of a transcriptional target-based approach to identifying morphogenetic genes, and suggest that br regulates morphogenesis through the regulation of genes involved in multiple critical processes.

Materials and MethodsFly stocks

All Drosophila stocks were maintained on media consisting of corn meal, sugar, yeast, and agar in incubators maintained at a constant temperature of 25° or at room temperature. w1118, ybr5, br1, distalless (dll)-Gal4, apterous (ap)-Gal4, P1, w1118, Pgm1LA00593, and the Transgenic RNAi Project lines (“short-hairpin” RNAi lines; https://fgr.hms.harvard.edu) were obtained from the Bloomington Drosophila Stock Center (Bloomington, IN). “Long-hairpin” RNAi lines were obtained from the Vienna Drosophila RNAi Center (Vienna, Austria; Dietzl et al. 2007). UAS-serp was obtained from Stefan Luschnig (University of Muster). UAS-Verm/CyO was obtained from Christos Samakovlis (Stockholm University). vkg-GFP (Flytrap; Buszczak et al. 2007) was obtained from Sally Horne-Badovinac (University of Chicago). w; hs-Z2 (CD5-4C); hs-Z2 (CD5-1) was obtained from Cindy Bayer (University of Central Florida). Loxl-1-RNAi (stock 11335R) was obtained from the National Institutes of Genetics Fly stock collection (Kyoto, Japan). dll-Gal4 was balanced with CyO, P (Le et al. 2006). w1118 was used as the wild-type control, unless otherwise noted. All fly stocks and reagents are listed and described in the reagents table.

br isoforms

To clarify the relationship between the isoforms of the br gene listed on FlyBase and the Br protein isoforms bearing the Z1, Z2, Z3, and Z4 zinc finger domains, translated sequences of all 15 br isoforms listed on FlyBase were downloaded from FlyBase and aligned using Clustal Omega (Goujon et al. 2010; Sievers et al. 2011; Thurmond et al. 2019). The first 431 amino acids represent the Br “core” region and were shared among all isoforms; the remaining amino acids were compared to the published Z1, Z2, Z3, and Z4 zinc finger domain sequences to determine which zinc finger was carried by each FlyBase isoform (DiBello et al. 1991; Bayer et al. 1996).

Fly staging, dissection, and photography of live leg imaginal discs

w1118 and ybr5/Binsn flies were staged on food supplemented with 0.05% bromophenol blue, as described in Andres and Thummel (1994). w1118 and ybr5/Y mutant animals were selected (mutant males were selected using the yellow marker) at −18 hr (blue gut larvae), −4 hr (white gut larvae), 0 hr (white prepupae), and +2 hr and +4 hr relative to puparium formation. Leg imaginal discs were dissected in phosphate-buffered saline (PBS), and then transferred to fresh PBS. Brightfield photomicrographs were captured within 5 min on a Nikon Eclipse 80i microscope equipped with a Photometrics CoolSNAP ES high performance digital CCD camera using a Plan APO ×10 (0.45 NA) objective. For the examination of collagen integrity in prepupal leg discs, we crossed ybr5/Binsn virgins to w1118, vkg-GFP males, and crossed F1ybr5/w1118; vkg-GFP/+ females to w1118 males. We selected ybr5/Y; vkg-GFP/+ and w1118/Y; vkg-GFP/+ white prepupae based upon the y phenotype and aged them at 25°. ybr5 prepupae were confirmed to contain the br5 allele based upon pupal morphology. We dissected leg imaginal discs at the indicate time points in PBS, mounted them live in mounting media (90% glycerol, 100 mM Tris pH 8.0, 0.5% n-propyl-gallate) and imaged them sequentially with brightfield microscopy and wide-field fluorescence microscopy on a Nikon Eclipse 80i microscope equipped with a Photometrics CoolSNAP ES high performance digital CCD camera, using a Plan APO ×20 (0.75 NA) objective. All images were captured with identical settings (light or fluorescence intensities, and exposure times). On average, three or four discs were imaged from each prepupa, and at least seven different prepupae were imaged at each time point. All digital images were cropped and adjusted for brightness and contrast in Adobe Photoshop (version CC 2018; Adobe, San Jose, CA) or ImageJ (version 1.51r; National Institutes of Health, Bethesda, MD), and figures were compiled using Adobe Illustrator (version CC 2018).

Trypsin experiments

w1118 and br5 white prepupae were dissected in PBS to isolate three leg imaginal discs from the same animal. These imaginal discs were incubated on 3-well depression slides in either PBS, PBS plus 0.025% trypsin (25200-056; Gibco) or PBS plus 0.0025% trypsin for 15 min at room temperature, at which point they were immediately imaged by brightfield microscopy on a Nikon Eclipse 80i microscope equipped with a Photometrics CoolSNAP ES high performance digital CCD camera using a Plan APO ×10 (0.45 NA) objective. In total, 21 w1118 and 17 br5 animals were dissected and imaged.

Immunostaining

Imaginal discs were hand dissected from w1118 or br5 +4 hr prepupae in fresh PBS, and fixed immediately in 4% paraformaldehyde for 20 min. The following antibodies were used at the given dilutions: rat anti-DE-cadherin (clone DCAD2 from Developmental Studies Hybridoma Bank at the University of Iowa) at 1:25, rabbit anti-Vermiform (gift from Stefan Luschnig, University of Munster) at 1:1000 (Luschnig et al. 2006), Donkey anti-rabbit Cy3 (Jackson ImmunoResearch Laboratories, West Grove, PA) at 1:400 and donkey anti-rat Cy2 (Jackson ImmunoResearch Laboratories) at 1:400. Confocal images were acquired on a Leica SPE laser scanning confocal microscope using an ACS APO ×40 (1.15 NA) oil immersion lens. All digital images were cropped and adjusted for brightness and contrast in ImageJ (version 1.51s; National Institutes of Health). Figures were compiled using Adobe Illustrator (version CC 2018).

RNA-sequencing and data analysis

To control for potential genetic effects from the autosomes and Y chromosome, we crossed ybr5/Binsn females to w1118 males and then crossed the resulting ybr5/w1118 females with w1118 males. This cross produced ybr5/Y and w1118/Y males that, at a population level, had identical autosomes and Y chromosomes. Since we selected the br5 animals based upon the yellow cuticular phenotypes, we tested to make sure that y and br did not recombine apart to any significant degree. y and br are reported to map 0.2 cM apart on the X chromosome (Gatti and Baker 1989), and in two separate experiments we did not detect any recombination between these genes (n > 200). After RNA-sequencing, we identified the br5 mutation as a C to T transition at position 1654571 in GenBank AE014298 (X chromosome of D. melanogaster), converting a conserved histidine in the zinc finger of the Z2 isoform into a tyrosine. We found that all the reads through this interval had the mutation in the br5 samples, whereas none of the reads from w1118 samples had the mutation.

Third instar ybr5/Y and w1118/Y larvae were staged on food supplemented with 0.05% bromophenol blue. Blue gut larvae (−18 hr) and white prepupae (0 hr) were selected and leg imaginal discs were hand-dissected in PBS. Triplicate independent samples were obtained for ybr5/Y and w1118/Y white prepupae and for w1118/Y −18 hr larvae. Total RNA was isolated using TriPure (Roche, Indianapolis, IN) and then purified over RNAeasy columns (Qiagen, Valencia, CA). Approximately 5 μg of total purified RNA was obtained for each sample, and ∼1 μg of each sample was provided to the Genome Sequencing Core at the University of Kansas for library preparation using the TruSeq stranded mRNA kit (Illumina, San Diego, CA). Single read 100 (SR100) was performed on a single lane of an Illumina HiSeq 2500 (Genome Sequencing Core, University of Kansas).

The quality of the raw RNA-sequencing reads was visually confirmed using FastQC (version 0.11.5; Andrews 2010), and both low-quality data and adaptor sequences were removed using Trimmomatic (version 0.36; Bolger et al. 2014). The number of remaining reads per sample averaged 18.1 million (range = 12.6–22.8), and all reads were at least 50-nt in length. Filtered reads were mapped to the D. melanogaster reference genome (Release 5.3) using TopHat (version 2.1.1; Trapnell et al. 2009; Kim et al. 2013). Default parameters were employed, with the addition of the “–no-novel-juncs” flag to use the gene annotation as provided, and the “–library-type fr-firststrand” flag since the RNA-sequencing data derives from the Illumina TruSeq stranded mRNA kit. On average 86.2% of the reads mapped to the reference across samples (range = 83.1–88.5). Genotypes were compared, and differentially-expressed genes were identified using the Cufflinks pipeline (version 2.2.1; Trapnell et al. 2010; Roberts et al. 2011; Trapnell et al. 2013). Specifically, we employed the “cuffquant” and “cuffdiff” routines using default parameters, adding the “-b” flag to run a bias correction algorithm that improves expression estimates (Roberts et al. 2011).

Bioinformatic analyses

Gene ontology analysis was performed using the Gene Ontology Consortium’s (http://www.geneontology.org) gene ontology enrichment analysis tool (Gene Ontology Consortium 2015). To ensure we were considering genes showing large-scale induction or repression, significantly differentially expressed genes showing ≥1.5-fold change between w1118 0 hr and br5 0 hr samples and at least 5 fragments per kilobase of exon per million fragments mapped (FPKM) in the sample showing higher expression (w1118 for br-induced genes, br5 for br-repressed genes) were used as input into the tool. In a second experiment, significantly differentially expressed genes showing ≥1.5-fold change between w1118 –18 and 0 hr samples and at least 5 FPKM in the sample showing higher expression (w1118 0 hr for developmentally induced genes and w1118 –18 hr for developmentally repressed genes) were used. The same data sets were used for the Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis using WebGestalt’s (http://www.webgestalt.org) overrepresentation analysis (Wang et al. 2017).

ATP measurement

Leg and wing imaginal discs were dissected from +4 hr w1118 and br5 mutant prepupae in PBS and immediately homogenized in 100 μl of lysis buffer (6 M guanidine HCl, 100 mM Tris pH 7.8, 4 mM EDTA) on ice and frozen. A mixed leg and wing sample was used in this experiment to allow all of the dissections to be completed in one sitting with enough material for the subsequent analysis. We note that prior Northern blot analysis using a mixed sample of leg and wing imaginal disc material showed identical expression of ecdysone and Rho signaling pathway gene expression to those composed only of leg imaginal disc material, and that wing discs arrest development in br5 prepupa at the same stage that leg discs do (R. Ward, unpublished data). After all of the samples were collected, they were thawed to 4°, and a 10 μl aliquot was taken for protein measurement using a Bradford Assay (Bio-Rad, Hercules, CA). The remaining sample was boiled for 5 min, spun at 13,000 rpm in a refrigerated microcentrifuge for 3 min, and then 10 μl of the supernatant was diluted into 90 μl of dilution buffer (25 mM Tris pH 7.8, 100 μM EDTA). This sample was further diluted 10-fold in dilution buffer and 10 μl of this sample was used for ATP quantification using the ATP Determination Kit (A22066; Molecular Probes, Eugene, OR) according to the manufacturer’s protocol, using a BioTek Synergy HT plate reader. Each biological sample contained ∼13 wing discs and 35 leg discs. Triplicate samples were processed and three technical replicates of each biological sample was assayed, along with a dilution series of ATP from 1 nM to 1 μM. Statistical analysis was performed using a type 3 ANOVA with protein level and treatment as factors.

Northern blot analysis

Progeny from a cross of ybr5/Binsn X Binsn/Y were staged on standard Drosophila media supplemented with 0.05% bromophenol blue, as described in Andres and Thummel (1994). Total RNA was isolated by direct phenol extraction from leg imaginal discs dissected from staged Binsn/Y males. Approximately 9 μg of total RNA per sample were separated by formaldehyde agarose gel electrophoresis and transferred to a nylon membrane. The membrane was hybridized and stripped as described in Karim and Thummel (1991). Generation of probe fragments for br (br core and br-Z2) and rp49 is described in Andres and Thummel (1994). Specific probes were labeled by random priming of gel-purified fragments (Stratagene, St. Louis, MO).

Functional analyses

To amplify potential RNAi in long-hairpin lines, we crossed a UAS-Dicer transgene onto the Gal4 lines used to drive the expression of the gene-specific double-stranded RNA to produce P1, w1118; Dll-GAL4/CyO, dfd-YFP and P1, w1118; ap-GAL4/CyO, dfd-YFP. For consistency, we also used these recombined lines when crossing to short-hairpin RNAi lines. Virgin females of these genotypes were then mated to UAS-RNAi males. The vials were kept in incubators maintained at a constant temperature of 25°. The adults were transferred twice to new vials, and newly eclosing F1 flies were separated by phenotype and examined for malformed third legs each day for a total of 8 days per vial. We considered an animal to be malformed if it displayed malformation in at least one leg, and defined a leg as malformed if any femur, tibia, or tarsal segment was bent, twisted, missing, or was excessively short and fat.

To control for background effects that may contribute to the appearance of malformed legs, we crossed UAS-Dcr1, w1118; Dll-GAL4/CyO, dfd-YFP virgin females with males from the w1118 line and a loxl1-RNAi line and examined the offspring for malformed legs. Curly-winged offspring from RNAi crosses, which carry the UAS-RNAi hairpin construct, but not the GAL4 driver, were also examined. A few of the RNAi constructs were carried over a balancer chromosome; only flies without the balancer were counted. Flies carrying malformed legs did not exceed 1% of all flies in either control cross, and exceeded 1% among curly-winged offspring from only two RNAi crosses. In neither of these two crosses (carrying RNAi hairpin constructs against CG7447 and E[spl]m4) did flies carrying malformed legs exceed 2% of all curly-winged flies, and neither of these crosses were among those in which a significant number of straight-winged flies carried malformed legs. For crosses in which we examined preadult stages expressing RNAi, individuals carrying the Dll-GAL4 driver were identified by the lack of dfd-YFP expression.

br-Z2 overexpression studies

Approximately 50 HS-br-Z2 or w1118 late larvae were placed into an empty fly vial with a piece of moist Whatman paper in the bottom of the vial. The vial was heat shocked in a 37° incubator for 60 min, after which any animals that had pupariated were removed. The vial was moved to a 25° incubator for 6 additional hours, at which point all prepupae were removed to a food vial for further development (also at 25°). After a further 18 hr, the prepupae were scored to determine if they had pupated. The animals were then left at 25° and scored for eclosing 4 days later. In some experiments, HS-br-Z2 or w1118 were collected as they pupariated, aged 4 hr at 25°, and dissected to examine their leg imaginal discs. Another subset was dissected at approximately 16 hr after pupariating to examine the terminal leg imaginal disc phenotypes. All dissections, microscopy, and image preparation were conducted as described above.

Adult specimen preparations

Adult legs were dissected from the third thoracic segment in PBS, cleared in 10% KOH overnight, and mounted in Euparal (Bioquip, Gardena, CA) on microscope slides. Images of adult leg cuticles were captured on a Photometrics CoolSNAP ES high performance digital CCD camera with a Nikon Eclipse 80i microscope. All digital images were cropped and adjusted for brightness and contrast in ImageJ (version 1.51r; National Institutes of Health) and figures were compiled using Adobe Illustrator (version CC 2018).

Data availability

Fly stocks are available upon request. Figure S1shows the gene structure and RNA isoforms of the broad locus. Figure S2 shows the results of luciferase ATP assays to quantify ATP levels between w1118 and br5 +4 hr leg discs. Figure S3 compares +4 hr leg imaginal discs from w1118, br5, and Dll > br-RNAi-expressing animals. Table S1 shows the genes that are differentially expressed between w1118 and br5 0 hr leg discs. Table S2 shows the gene ontology terms that are significantly over- and underrepresented in the br-induced and br-repressed gene sets. Table S3 shows the KEGG pathway analyses on the br-induced and br-repressed gene sets. Table S4 shows the genes that are differentially expressed between −18 and 0 hr in w1118 leg discs. Table S5 shows the gene ontology terms that are significantly over- and underrepresented in the w1118 developmentally regulated gene sets. The reagents table lists all the stocks and reagents used in this study. All RNA-sequencing data sets are available from the Gene Expression Omnibus. The project accession number is GSE140248, and the project title is “Genome-wide differences in RNA expression in Drosophila melanogaster leg imaginal discs based on time and presence/absence of broad-based gene regulation.” Supplemental material available at figshare: https://doi.org/10.25386/genetics.11729019.

ResultsPhenotypic characterization of br5 mutant leg imaginal discs

To identify genes regulated by br during metamorphosis using an RNA-sequencing approach, we wanted to use a null br allele. The br5 allele has been reported to be amorphic, with homozygous mutants failing to develop past the early prepupal stage (Kiss et al. 1988). Consistent with this notion, br5 encodes a protein with a His492–Tyr substitution in the conserved Z2 zinc finger domain, suggesting that it has defective DNA binding properties (L. von Kalm, personal communication; confirmed in the RNA-sequencing; see Materials and Methods). We previously used live imaging to show that br5 hemizygous mutants failed to complete metamorphosis (Ward et al. 2003). To more closely examine how imaginal disc development is disrupted in these animals, we dissected and compared leg discs from br5 and w1118 larvae and prepupae (Figure 1). br5 leg discs resemble w1118 leg discs during larval stages (through 4 hr before pupariation), consisting of a columnar epithelium with folds that make three to four concentric circles. The discs bulge from the central circles (which develop into the distal-most leg segments), and are covered by the peripodial epithelium. By the prepupal stage, leg discs begin to differ between the two genotypes. Both br5 and w1118 discs begin to elongate in a telescoping fashion at the onset of metamorphosis (0 hr), beginning with the centermost regions. However, by 2 hr after pupariation, elongation of br5 discs clearly lags behind that of w1118 discs. w1118 discs continue to elongate and by 4 hr after pupariation, the future five tarsal segments and distal tibia are clearly identifiable underneath the peripodial epithelium. br5 mutant discs, on the other hand, form shorter, wider structures with deeper folds between the tarsal segments. br5 mutant discs show only limited elongation over subsequent hours and fail to evert to the outside of the prepupa. This phenotype is completely penetrant (n > 1000 leg discs examined). Notably, imaginal discs can be found inside degenerating br5 mutant late prepupae that look similar to +4 hr mutant discs, including having an intact peripodial epithelium (n > 20 animals; Figure 1K).

Figure 1Figure 1Figure 1

Broad is required for normal prepupal leg imaginal disc development. Brightfield photomicrographs of leg imaginal discs from w1118 (A–E) and br5 (F–K) mutant larvae and prepupae. Times given are relative to puparium formation. Note that leg imaginal discs are similar between w1118 and br5 during larval time points (−18 and −4 hr), but are noticeably different starting at 0 hr. Although w1118 discs elongate from the center (distal tarsal segment; indicated by asterisk) and are segmented by +4 hr, br5 discs show limited elongation and have wider tarsal segments. Intact leg imaginal discs can be found inside dead br5 prepupae as late as 17 hr after pupariation. These late imaginal discs have not elongated much past the +4 stage and often have an intact peripodial epithelium (arrow). Bar, 100 μm.

The disruption of prepupal elongation in br5 mutants, coupled with the presence of the peripodial epithelium in late-staged prepupae (which is known to be degraded by matrix metalloproteinases; Proag et al. 2019), motivated us to examine ECM breakdown in wild-type and br5 mutant discs. We therefore crossed a GFP-tagged version of Collagen IV (Vkg-GFP) into w1118 and br5 and examined their leg discs at various time points during metamorphosis (Figure 2). At early stages (0 and 2 hr after pupariation), wild-type and br5 leg discs closely resemble one another, with fibrous cylindrical or nearly conical basal ECM structures. By 4 hr after pupariation, some wild-type discs exhibit degradation of the basal ECM. This degradation takes the form of a “clearing” of a central channel along the length of the disc. In total, 37.5% of wild-type discs (n = 32, from 12 animals) showed this clearing at 4 hr after pupariation, 66.7% (n = 21, from 12 animals) at 6 hr, and 100% (n = 12, from 8 animals) at 7 hr (Figure 2K). In contrast, most br5 mutant discs retained the fibrous ECM appearance as late as 8 hr after pupariation (Figure 2, J and K). Only 12.0% of discs (n = 25, from 7 animals) showed partial clearing; the other discs showed no clearing.

Figure 2Figure 2Figure 2

broad is required for efficient degradation of the basal ECM in prepupal leg imaginal discs. Wide-field fluorescence (A–J) and brightfield (A’–J’) photomicrographs (at the same focal plane) of leg imaginal discs from w1118 (A–E) and br5 mutant (F–J) prepupae expressing Viking-GFP (Collagen IV). Ages of the animals relative to puparium formation are indicated above the figures. Note that Viking-GFP forms cables of ECM lining the lumen at the basal side of elongating leg discs, which are gradually cleared from +4 to +7 hr APF in w1118 leg discs (white arrows in C, D, E, and J; black arrows in C’, D’, E’, and J’ represent matching location in brightfield images). Similar clearing is not observed in the br5 mutant discs and persists at least through 8 hr APF (J) in many animals. Bar, 100 μm. (K) Graph showing percentage of discs showing partial or total clearing of GFP expression in the channel.

Since br5 mutant leg imaginal discs show reduced elongation and defective proteolysis of the ECM, we wondered whether an exogenous protease could restore normal elongation to the br5 discs. We therefore dissected three leg imaginal discs from individual w1118 or br5 animals at the onset of metamorphosis and treated one with 0% trypsin (PBS control), another with 0.0025% trypsin, and the third with 0.025% trypsin for 15 min at room temperature. Leg imaginal discs from 18 of the 21 w1118 animals had clearly elongated after 15 min in 0.0025% trypsin (Figure 3B). The tarsal segments in these discs were still obviously segmented, although in some cases the depth of the folds at these segment boundaries was reduced relative to the PBS control discs. w1118 leg discs incubated for 15 min in 0.025% trypsin showed even greater elongation (in 100% of the discs), but only with noticeable tarsal segment boundaries in two of the discs, whereas the remainder had elongated discs with the tarsal segments appearing as one long continuous segment (Figure 3C). In the br5 mutant leg discs, there was a clear difference in the morphology of the leg discs at the onset of the experiment (which remained unchanged through 15 min in PBS; data not shown) compared to the w1118 discs, in which the distal tarsal segment was rounder and the folds between the tarsal segments were noticeably deeper (Figure 3D vs. Figure 3A). A total of 10 of the 17 discs treated with 0.0025% trypsin showed some elongation of the tarsal segments, but no shallowing of the segment boundaries (Figure 3, E and G). Similarly, 12 of the 17 discs treated with 0.025% trypsin showed an increase in the degree of elongation (Figure 3G). The morphology of these discs was clearly distinct from w1118 treated similarly. The absolute level of elongation in br5 was less than in w1118, and more interestingly, the br5 discs did not eliminate the folds between segments, but rather deepened those folds (Figure 3F). This morphology was observed in 14 of the 17 discs examined (Figure 3G).

Figure 3Figure 3Figure 3

br5 leg discs show aberrant elongation upon application of trypsin, and retain anisometric cell

留言 (0)

沒有登入
gif